Download document 10502369

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Mercury-arc valve wikipedia , lookup

Electric power system wikipedia , lookup

Switch wikipedia , lookup

Induction motor wikipedia , lookup

Ground (electricity) wikipedia , lookup

Transformer wikipedia , lookup

Ohm's law wikipedia , lookup

Islanding wikipedia , lookup

Power engineering wikipedia , lookup

Pulse-width modulation wikipedia , lookup

Brushed DC electric motor wikipedia , lookup

Electrical ballast wikipedia , lookup

Resistive opto-isolator wikipedia , lookup

Current source wikipedia , lookup

P–n diode wikipedia , lookup

Power inverter wikipedia , lookup

Capacitor wikipedia , lookup

Electrical substation wikipedia , lookup

Power MOSFET wikipedia , lookup

History of electric power transmission wikipedia , lookup

Three-phase electric power wikipedia , lookup

Stepper motor wikipedia , lookup

Distribution management system wikipedia , lookup

Stray voltage wikipedia , lookup

Transformer types wikipedia , lookup

Schmitt trigger wikipedia , lookup

Voltage regulator wikipedia , lookup

Variable-frequency drive wikipedia , lookup

Surge protector wikipedia , lookup

Power electronics wikipedia , lookup

Opto-isolator wikipedia , lookup

Rectifier wikipedia , lookup

Voltage optimisation wikipedia , lookup

Alternating current wikipedia , lookup

Buck converter wikipedia , lookup

Spark-gap transmitter wikipedia , lookup

Mains electricity wikipedia , lookup

Switched-mode power supply wikipedia , lookup

Transcript
http://www.modelengineeringwebsite.com/EDM_1.html
ON THE long journey back to Sussex from the Harrogate Exhibition 2008, Alan Wragg and I were
chatting about the highlights of the show. Maurice Fagg’s ‘Simple Spark Eroder’ (his words)
attracted a great deal of attention, and we resolved to find out more about spark erosion, or to
give it its correct title, Electrical Discharge Machining; EDM for short.
This proved unexpectedly difficult. The internet was the obvious place to start, but although there
are many sites extolling the virtues of EDM and offering machines for sale, technical information
was practically non-existent. We had some luck in finding a book written by Ben Fleming, an
American, which described how to build one, but the content was rather light on theory. However,
slowly, from many sources, enough information was gathered to allow us to have a go at
designing and building our own machine.
This was the MKAW mk1 (Mike Kapp Alan Wragg mark 1), as seen at several exhibitions during
2008 and early 2009. The main feature of this unit was its configurability. We still did not fully
understand the underlying principles, so most parameters were adjustable. We could alter
voltages, set charging currents, and select circuit capacitance. This unit taught us much, the most
important of which was that none of the above settings seemed to matter much, but that
mechanical integrity mattered a great deal.
So how do you make a simple spark eroder?
The essential building blocks are:
1.Power supply
2.Pulse generator
3.Mechanical system, comprising:
Tool holder, Work-piece clamp, Tool/work-piece Gap control mechanism, Swarf removal.
And for hands-free operation:
4.A Servo System to automatically control the gap size
Using the knowledge gained from the mark 1 and subsequent experiments, we would like to offer
you what we believe is the world’s simplest spark eroder as shown in the block diagram.
On the right is a container full of a dielectric. In this context, a dielectric is a liquid that is free of
ions and, therefore, does not conduct electricity. In the beginning we used paraffin, but now we
mostly use naturally precipitated dihydrogen monoxide, (more commonly known as rainwater).
Tap water DOES NOT work as it contains dissolved salts, a rich source of ions. Rainwater is
essentially distilled water, so contains no salts, and it’s free. (In the dry season you can get deionized water from Halfords, at a price.)
Immersed in the dielectric are the work piece and the cutting tool. There are no forces acting on
the work piece other than the circulating liquid, so it can be held in place relatively lightly. The
cutting tool is generally made of copper, brass, or graphite, regardless of how hard the work piece
is. Copper will happily cut tungsten carbide or hardened tool steel, although the tool will also
erode.
This is how it works
If two conductors (tool and work piece in our case) have a voltage across them and approach
each other, a point is reached where the electrical field gets so intense that it causes the
dielectric to ionize and conduct electricity. This is before the tool touches the work piece. A high
current flows across this tiny gap, melting a small area of the work piece. At the same time the
liquid boils, causing a small explosion which blasts the molten metal away, thus leaving a tiny
crater. Repeat thousands of times a second and you have a very effective cutting action.
The result is metal removal with no force applied, so it’s possible to machine fragile or very brittle
materials, given that they conduct electricity. You can cross existing holes at slight angles,
because there is no tendency
to follow existing holes. You can cut holes of any shape you can produce in easily machined
materials, such as graphite or brass. This makes it perfect for removing broken taps or drills, with
no damage whatsoever to the work piece. Just make a tool with a diameter slightly larger than the
core of the tap (base of the flutes) but smaller than the thread. Remove the core of the tap by
spark erosion and the remaining bits of the tap will fall out.
The main down sides of EDM are that you have to work under a liquid, and it is a slow process.
It is essential that a small gap is maintained between tool and work piece, small enough for
sparks to occur but not touching, so backlash-free fine mechanical control is an important
requirement.
How to build one
If you only want to build the device and not worry about all the calculations, just obtain the
components listed at the end of this article.
If you would like to design your own, perhaps a more powerful version, a DVD is available which
goes into detail about component selection calculations and shows how to assemble the
machine.
Referring to the block diagram, the first requirement is a source of direct current. The DC voltage
we recommend is based on it being high enough to spark across a reasonable gap, say .001in.,
but low enough not to be dangerous. Remember that the voltage is connected across the tool and
work piece, therefore, freely available to be touched. We have successfully used voltages
between 40 and 75 volts.
The Power Supply
Let us settle on a nominal 50 volts. To obtain this from our 230v RMS mains supply we need to
use a transformer. This is essentially a device which magnetically couples two or more coils of
insulated copper wire wound around a common iron core. The mains supply is connected to the
primary coil and the desired voltage appears across one or more secondary coils. Note that the
secondary coils are electrically isolated from the mains supply, making the unit relatively safe in
operation. You have to buy a suitable transformer and therefore need to know what to ask for.
Alternating Current (AC) starts at zero volts, builds to a peak, decays back to zero and then
repeats this cycle in the opposite direction, so its average voltage or current is zero, rather
inconvenient when asked what the voltage is!
However, while the current is flowing, no matter which way, it is doing work. It will heat
up a kettle element just fine, but not as fast as a steady Direct Current (DC) of the same
voltage as the AC peak. In other words, if a DC voltage boils some water in 1 minute, it
will require an AC voltage with a higher peak value to do the same – 1.414 times higher
in fact. (This number is the Square Root of 2.)
When turning AC into smooth DC you will find that the resulting DC voltage is the same
as the peak of the AC voltage. The transformer therefore needs to output 50v peak, so
ask for a transformer with (50 divided by 1.414) = 35 volts output.
In practice, off the shelf transformers often come with 2 identical secondary windings
which will be connect in series, so our transformer has 2 x 18 volt outputs. You also have
to specify the current rating of the transformer. This should be 3 amps or more, but note
that higher currents come at higher prices. The transformer we used is rated at 3.3
amps.
Connect a power lead and fused mains plug to the primary winding of the transformer
(see label on transformer for winding identification.) INSULATE the connections
THOROUGHLY. Under no circumstances must there be any place where it is remotely
possible to touch mains bearing metal. Fit a low amperage fuse – 3 amps maximum – to
the plug. Restrain the cable by tying it to the transformer with string. (Not wire!)
The 2 secondary windings are connected in series by connecting the ‘bottom’ of one
winding to ‘top’ of the other. Again refer to the transformer label to identify the windings.
Insulate this connection. The remaining 2 loose ends are fitted with ¼in. female spade
connectors.
Photo 1 shows the label on the transformer we used.
the corner is chamfered, denoting the positive DC output terminal. The one diagonally
across (furthest from the camera) is the negative DC output The transformer is
connected across the remaining two terminals, and being AC, it doesn’t matter which
way round.
To complete our power supply we need one more function, that of smoothing the rectified
pulses into proper DC. This is achieved by using a reservoir capacitor.
A capacitor consists of two conducting surfaces separated by an insulator. The unit of
capacitance is the Farad, the size of capacitor which will charge to one volt when one
amp flows into it for one second. Although this is a straightforward relationship, the unit is
very large. The most commonly used unit is the MICROFARAD (uF), i.e. one millionth of
a farad. Capacity is dependent on the area of the plates, the nature of the insulator and
its thickness. For high capacity we need large plates and thin insulation. Electrolytic
capacitors offer high capacity in a small part. The electrolytic capacitor used is polarized,
and this is indicated by marking the negative terminal. If connected the wrong way round,
the electrolyte will overheat and the capacitor will explode! A capacitance of a couple of
thousand microfarad will work well – we used 2,200 uF. Note that it must have a working
voltage somewhat higher than the maximum circuit voltage, so ours is rated at 63
The capacitor is fitted across the + and – terminals of the rectifier. If you now plug the
unit in and measure the voltage across the capacitor you should find it to be in the
vicinity of 55 volts. This is higher than calculated, for two reasons: 1, we are using 36
volts rather than the calculated 35 volts, and 2, transformers are rated at full load, but
there is no load on it at the moment.
This completes the power supply of our EDM.
Next time we look at the pulse generator.
By connecting the black and yellow wires together, we get nominally 36 volts from the
red and orange wires.
So, having chosen a transformer, what’s next? The output of the transformer is still AC,
so it has to be converted to DC. To do this we use DIODES. A diode is an electronic
device which conducts electricity in one direction only. The accepted method requires 4
diodes. These are usually encapsulated into a block with 4 terminals, known as a full
wave rectifier. Photo 2 shows the one we used, attached to an aluminium heatsink made
from a piece of square tube.
reduce the resistor to 12 ohms 100 watts, but it must be mounted on a heat sink, and shorts must be quickly
cleared.
Capacitor values between about 3 microfarad to several hundred microfarad will work. Small capacitors fire
frequently but with little power, whereas large ones fire less frequently but with fatter sparks. Around 30 to 100
microfarad is a good compromise. These should be NON-POLARISED capacitors of at least 100 volts working
ratings. Photo 4 shows the completed RC circuit, with 2 capacitors fitted. Note the two aluminium rails, or ‘busbars’, which allow the number of capacitors to be varied. Feel free to experiment with different numbers of
capacitors.
Finally, we need some indication of the gap size to guide the operator. By connecting a LED across the tool and
work-piece, it will be:
--bright when the gap is too large
--dim when sparking properly
--out when the gap is too small
Next time we will look at the mechanical parts of the machine.
To see part one, click here
Go to part three, click here
Part four here
The Pulse Generator
Over the years two practical methods of forming pulsating DC have emerged, the Resistance/Capacitance (R/C)
oscillator, and Power Switching. We will be using the R/C circuit. The R/C oscillator is very simple and was in
common use before high power transistors made power switching possible.
The output of the power supply charges a capacitor via a resistor. This causes the voltage across the capacitor to
ramp up over time. This same voltage also appears across the tool to work-piece gap, and if the gap is small
enough a spark will jump the gap. Because there is little resistance between the capacitor and the tool/workpiece, a heavy current, perhaps hundreds of amps, will momentarily flow through the ionised dielectric. The
capacitor discharges to a low voltage, causing the spark to go out, so the capacitor can now charge up again. The
whole process repeats until the gap becomes too big for the spark to strike. Some mechanism must be provided
to control the gap size. This is either the Mk1 human hand, or for the more adventurous, a servo motor.
The size of resistor and capacitor is determined by the power available and the cutting action preferred. For
roughing, fat (high current) sparks are best, while for finishing lower current sparks applied more frequently work
best.
A bigger capacitor takes longer to charge up but gives a fatter spark. A lower value resistor charges the capacitor
more quickly, but puts a greater strain on the power supply. We have a nominal 50v power supply with a
maximum 3 amp output. It is unwise to continuously draw more than 2/3rds of the maximum available current,
which is 2 amps. With the tool shorted to the work piece the resistor will have to absorb this power, so
R=voltage/current = 50/2 = 25 Ohms.
HOWEVER - We do not draw power continuously, and the AVERAGE power drawn is much less than the
calculated worst case. If we make sure shorts never last long, we can use lower resistance values and,therefore,
lose less power just heating up the resistor.
If we halve the calculated resistor value to say 12 ohms, then 50 volts/12 Ohms = 4+ Amp peak, which is 50v x
4A = 200 Watts. However, this only applies if the tool shorts to the work-piece; in normal operation the current
only flows intermittently. This allows us to
Y
OU NEED to remember that the tool to work piece gap is extremely small, often less than 0.001in. so
what is required is a method of feeding the tool in the Z plane only, very slowly and accurately. As little as 0.001in.
of backlash or sideways movement can render the machine useless. We use spring and/or gravity biasing to
eliminate backlash. To prevent sideways movement the roller bearings are spring biased by slightly twisting the
plastic mounting plate, while backlash in the Z-plane is eliminated by ‘hanging’ the entire slide mechanism from a
3mm threaded rod.
Those of you who like to devise you own methods, feel free to do so, and if you find an accurate and easy to
construct method, please let us know.
This is what we did.
Everything is mounted on a piece of white coated chipboard. The support column is a 16mm diameter rigid tube
firmly attached to the baseboard. The part of the ‘fork’ touching the column is 15mm thick, and the two ‘tines’
attached to it are about 100mm by 15mm, and 6mm thick. The two guide rods are 6mm precision ground silver
steel, accurately parallel, passing through the fork tines and held in place with set screws. The plastic tool carrier
is 6mm thick, and the four rollers are attached so that they ride firmly up against the guide rods. The rollers
themselves are roller bearings inside an aluminium sleeve, the sleeves recessed so that they part-wrap onto the
guides. The result is a slide that freely moves up and down, but has zero play in any other direction – this is very
important.
Attached to the bottom of the tool holder is an aluminium block which incorporates a nylon insert threaded M3.
The leadscrew screws into this insert, thereby electrically insulating the block from the leadscrew. The entire slide
hangs on the leadscrew, thereby eliminating backlash; the smallest movement of the operating wheel will slightly
lift or lower the tool. There are passages drilled in the tool carrier block which allows dielectric liquid to be sucked
up through the tool, into the carrier block, and out through a tube attached to it. This is useful when machining
deep holes, as it removes swarf from the bottom of the hole before it can cause problems.
We incorporated a clamp rod which both holds the work-piece in place and makes the electrical connection to it. It
also serves as additional support for the slide head, making the whole assembly much more rigid and trouble-free
in operation.
Electrical connections
The tool carrier block is fitted with a push-on male spade connector. This is wired to the negative side of the 33uF
capacitor(s), as close as possible to the capacitors so as not to cause high currents to flow through other parts of
the circuit.
The positive side of the 33uF capacitor(s) is wired to the clamp rod, or to the support column if you prefer.
PARTS:
We leave the sourcing and shaping of mechanical parts to you, but here are the details of the electrical parts.
They were all obtained from RS components, total cost being in the region of £35.
Transformer, 2 x 18volt, 3.3 amp;223-8033£16
Bridge Rectifier, 15A, 200V;629-6689£2.75
Capacitor, 2200uF, 63 Volt208-7245£3
Resistor, 12 Ohm, 100 Watt;252-2912£4.50
Capacitor, 33uF, 100 Volt, non-polarised521-3475£3.50 (5)
The costs quoted are for guidance, VAT to be added.
A
LTHOUGH manual operation is surprisingly easy, twiddling the knob for a deep hole can be
tedious. By fitting a motor and some simple electronics the unit can be left to get on with it while you
do something else.
Once again, the design criteria was dominated by the desire to make the device accessible to those
not electronically gifted. So, no printed circuit board, no microprocessor (therefore no software), and
(almost) no soldering.
We used a readily available 12 volt 60 RPM geared motor in the prototype. This was coupled to the
vertical (Z-axis) screw using a 4:1 reduction belt drive.
Physical construction of electronic circuits are often a problem for those without electronic skills. If you
have the skill, go ahead and make yourself a printed circuit board. If not, we suggest you buy a
prototyping board, which allows you to assemble an electronic circuit without even a soldering iron!
For those in the know, the circuit looks like the diagram on the right.
How the circuit works
If you don’t want to know how the circuit works, you can skip to ‘Construction’, otherwise read on from
here.
To power the motor and its associated electronics we need 12 volts, to be derived from the main 50
volt supply. A conventional linear regulator would not be comfortable with such a large voltage
difference, and would also waste power. A small, efficient 12 volt switching regulator is, therefore,
used instead. The only unusual feature of the circuit is the 22v Zener diode at the input; this is to
reduce the voltage into the regulator to below 40 volts, the highest the chosen regulator can handle.
The control circuit is built around a 14 pin CD40106 Hex Schmitt Trigger chip, but only four of its six
identical elements are actually used. In the diagram they are numbered from 1 to 4 for reference
purposes. A feature of the device is that the output of each trigger is at either 0v or 12v; never
between these values. If the input voltage slowly rises from 0v to below about 6.3v, the output remains
at 12v. As the input rises a bit more the circuit will suddenly ‘trigger’ and its output will instantly drop to
0v. Conversely, if the input slowly reduces from 12v to about 5.7v the output remains at 0v, but if it
goes a little below this the trigger action will cause the output to suddenly switch
to 12v.
The control circuit monitors the voltage across the spark gap, between tool and workpiece, and runs
the motor accordingly. If the gap voltage rises above about 36v the gap is getting too large, so the
motor has to turn the screw counter-clockwise to reduce the gap. Conversely, if the voltage does NOT
rise above about 36v the motor needs to turn the screw clockwise so as to increase the gap.
Voltage sensing operates as follows (refer to the diagram above):
Assume the tool is well above the workpiece, so that the voltage on the workpiece rises to the full 50plus volts. The 24v Zener diode subtracts 24 from the 50v so 26v is applied to the top resistor of the
pair connected between the Zener and ground (Note that the tool is also connected to ground). The
two resistors are of equal value, so the voltage divides equally between them, leaving their mid-point
trying to rise to 13v. If we allowed this to happen the CD40106 chip could be damaged, but by
connecting a 10v Zener between this point and ground the voltage cannot rise above 10v, so the
danger is eliminated.
So, with the input of trigger 1 at 10v its output will be at 0v. This is the condition before sparking starts,
with a big gap between tool and work-piece. The output of trigger 1 connects to both triggers 2 and 4,
so with trigger 1 at 0v, the output of 2 and 4 will be at 12v. Trigger 2 connects to trigger 3, so the
output of 3 will be at 0v. This is important – the extra stage of inversion provided by trigger 3 means it
is impossible for the outputs of triggers 3 and 4 to ever be at the same level, so we will never attempt
to drive the motor in both directions at the same time!
To summarise, the output of trigger 3 is at 0v, while that of trigger 4 is at 12v. Trigger 3 output
connects to the top left and bottom right MOSFET switches, holding them open. Trigger 4 connects to
the bottom left and top right switches, holding them closed. Current can therefore flow from the 12v
supply at the top through the top right switch, through the motor from right to left, then to ground
through the bottom left switch. This causes the motor to turn clockwise so lowering the tool towards
the work-piece.
When the gap gets small enough a spark will occur, causing the voltage on the work-piece to drop to
about 12 volts, at which point the spark will go out. Consequently the voltage at the input of trigger 1
will drop to zero causing the voltages on all four triggers to reverse. The voltage across the motor will
also reverse, but only very briefly – for perhaps a thousandth of a second. This is because the voltage
on the work-piece will start to rise again as soon as the spark goes out, and when it gets to about 36
volts all the triggers will again reverse. The motor, therefore, experiences a series of pulses which
reverse in polarity every time a spark occurs. If the gap is good the positive and negative pulses will be
of roughly equal length but too short for the sluggish motor to respond to. However, as the gap grows
larger the voltage on the workpiece will spend more time above 36v than below it, so the ‘motor-down’
pulses will become longer than the ‘motor-up’ pulses, causing the motor to lower the tool. The reverse
is also true; if the gap becomes too small, as when swarf builds up in the gap, the spark will occur at a
lower voltage, causing the ‘motor-up’ pulses to dominate, resulting in the tool being lifted. (This is
technically known as Pulse Width Modulation, or PWM for short.)
Although this circuit works well, it lacks facilities for initial set-up. With a heavily geared motor it is
impossible to manually turn the Z-axis screw in order to set the initial tool height, and the moment you
turn the power on the motor will start winding the screw. To facilitate set-up, two switches have been
added; a Stop/Start switch and a motor Up/Down switch. The Stop switch applies 12 volts, via two
diodes, to the inputs of triggers 3 and 4, thereby locking their outputs low and opening all four motor
control MOSFET switches, so no current flows through the motor. At the same time it supplies 12
volts to the centre of the 3-position manual motor control switch. This switch overrides the outputs of
triggers 3 and 4, so allowing the motor to drive he tool up or down at the behest of the operator. When
the Stop/Start switch is in the Start position the 12 volts is removed from both the triggers and from the
Up/Down switch, so total control is back with the voltage comparator circuit.
Electronic experts may wonder why we did not use one of the widely available motor control chips.
This was to allow the simple description of circuit operation given above, without having signals
disappearing into a ‘black box’. By all means use one if you wish. Half of a L293D (the one with
internal protection diodes) works perfectly well.
Construction
As mentioned earlier, construction consists of plugging components into sockets on a prototyping
board, such as the one shown right. This was obtained from RS Components under part number 488618, cost around £15.
There is not enough detail in the photograph to allow error-free construction, but the following
drawings should fill in the gaps.
First insert wire links as shown in the diagram. Use single strand 0.6mm insulated wire. Note that there
is no connection where wires cross.
Then fit components as shown in the lower diagram.
Test and set-up
If you have a volt meter it is wise to test as you go along. First fit components 1 to 6, connect to the 50
volt supply, and check that there is 12 volts across the top and bottom rows of sockets. Disconnect
from the 50 volt supply.
Now fit components 17 to 22, and wire up the motor and the up/down switch. Temporarily connect the
centre of the switch to 12 volts (top row of sockets) and the left hand sides of resistors 17 and 18 to 0
volts (bottom row of sockets). Connect to the 50 volt supply and operate the motor up/down switch.
Make sure that when the switch is up the tool rises and when it is down the tool descends. If it works
the wrong way round reverse the wires going to the motor – DO NOT reverse the switch. If all is well,
disconnect the 50 volts and remove the temporary wiring, remembering to restore the switch wiring to
the correct connections. Now fit the rest of the components, connect row 16 to the 33mfd capacitor as
shown, and you are ready to roll.
With the Stop/Start switch set to stop (pointing up) set the tool a millimetre or so above the work piece
using the Up/Down switch. With the electrolyte in place, switch to Start. The tool should descend fairly
rapidly until sparking starts, after which it should maintain the optimum gap for continuous sparking.
Component list
Note that RS part numbers are given below simply because that’s where I got most of the parts from.
For the cheaper components RS will only supply in multiple quantities. There are other suppliers who
may supply smaller or even single quantities. The RS web site is www.uk.rs-online.com.
1. 330 uH inductor RS 233-5241
2. Diode 11DQ06 RS 254-0702
3, 7: Zener Diode 24v RS 295-5154
4. Switching Regulator LM2574N-12 RS 533-3125
5. Capacitor 22mfd 63v RS 191-7905
6. Capacitor 220mfd 35v RS 526-1834
8, 11,12, 15, 16, 17, 18: Resistor 2.2K RS 149-739
9. Zener Diode 10v 812-465
10. CD40106 Hex Schmitt Trigger RS 571-1661
13, 14: Diodes 544-3480
19, 20, 21,22: MOSFETS VNP5N07 RS 313-3096